Bozzette S A, Kanouse D E, Berry S, Duan N
Health Sciences Program, RAND, Santa Monica, CA 90407-2138.
JAMA. 1995 Jan 25;273(4):295-301.
To evaluate the functional and health status implications of prescribing zalcitabine or zidovudine for initial therapy of acquired immunodeficiency syndrome (AIDS).
A substudy of a randomized controlled trial.
Private and public clinics and referral centers.
Had human immunodeficiency virus (HIV) infection, less than 0.20 x 10(9)/L (200 microliters) CD4+ cells, and either a history of Pneumocystis carinii pneumonia or symptoms of HIV infection. Fifty-eight percent (338/668) of main study enrollees representing 90% of enrollees at participating sites were included in this substudy.
Either zalcitabine at 0.75 mg every 8 hours plus inactive capsules identical in appearance to zidovudine or zidovudine at 200 mg (later 100 mg) every 4 hours plus inactive tablets identical in appearance to zalcitabine.
Results of a periodically completed self-report survey instrument containing specific questions about symptom impact, disability, work, functioning, and utilization as well as nine health and functioning scales adapted from the Medical Outcomes Study (MOS).
Zalcitabine recipients were twice as likely to undergo an invasive procedure (P = .004) or be admitted to hospital (P = .01). Zalcitabine recipients reported greater than 40% more symptoms that interfered with activity (P = .001) and greater than 50% more disability days (P < .01). They also had a 7% lower employment rate and a 35% lower monthly income. Average observed health status scores were lower in zalcitabine recipients overall, but especially in the early portion of the study. New methods for combining survival and health status data showed that, over 76 weeks of study, a typical zidovudine recipient spent about 4 (10%) more weeks with at least the typical health state than did a typical zalcitabine recipient.
Zidovudine has substantial advantages over zalcitabine in initial monotherapy of AIDS in terms of functional outcomes such as symptoms impact, disability, work, utilization, and health status. In this case, the differences in functional outcomes presaged differences in physiological and clinical measures. The inclusion of functional outcomes can greatly improve the information available from a clinical trial.
